Minimization of Surface Deflection in Rectangular Embossing Using Automatic Training of Artificial Neural Network and Genetic Algorithm

  • Sungmin Cho
  • Wanjin ChungEmail author


Surface deflection is a phenomenon that causes fine wrinkles on the outer surfaces of sheet metal and deteriorates product external appearance. It is quantitatively defined as the difference between the section curve of the sheet and the ideal curve. In this study, using neural networks, a prediction model for surface deflection according to material properties was constructed and combined with a genetic algorithm; the combination of the material properties was studied to predict the minimum surface deflection. Because of the limited number of simulation data, neural networks were developed using several sampling methods such as central composite design, Latin hypercube sampling, and random sampling. In the training of the neural networks, the optimal hyper-parameter of the neural network was found automatically using Latin hypercube sampling. In conclusion, for prediction of surface deflection in rectangular embossing, neural networks made by central composite design showed the best performance. In addition, it was confirmed that the procedure of combining automatic training of a neural network and the genetic algorithm accurately predicted the set of material properties that generates the minimum surface deflection. Also, the quantity of surface deflection predicted by the neural network was very close to that predicted by finite element analysis.

Key Words

Surface deflection Artificial neural network Finite element analysis Data sampling Genetic algorithm 



This study was supported by the Research Program funded by SeoulTech (Seoul National University of Science and Technology).


  1. Alam, F. M., McNaught, K. R. and Ringrose, T. J. (2004). A comparison of experimental designs in the development of a neural network simulation metamodel. Simulation Modelling Practice and Theory 12, 7–8, 559–578.CrossRefGoogle Scholar
  2. Box, G. E. P., Hunter, W. G. and Hunter, W. G. (2005). Statistics for Experiments. Wiley-Interscience. New York, USA.zbMATHGoogle Scholar
  3. Breitsprecher, T. and Wartzack, S. (2014). Neural network based modeling and optimization of deep drawing-extrusion combined process. J. Intelligent Manufacturing 25, 1, 77–84.CrossRefGoogle Scholar
  4. Cho, S. M., Son, D. G. and Chung, W. J. (2018). Minimization of surface deflection is rectangular embossing using machine learning. Proc. AEPA2018, Jeju, Korea.Google Scholar
  5. Chung, W. J., Chung, Y. C. and Kim, W. S. (2015). A robust prediction method of surface deflection using stoning simulation and curvature analysis. Steel Research Int. 86, 8, 886–893.CrossRefGoogle Scholar
  6. Glorot, X. and Bengio, Y. (2010). Understanding the difficulty of training deep feedforward neural networks. Proc. 13th Int. Conf. Artificial Intelligence and Statistics, Sardinia, Italy.Google Scholar
  7. Goldberg, E. D. (1989). Genetic Algorithms in Search, Optimization and Machine. Addison-Wesley Publishing Company. Boston, Massachusetts, USA.zbMATHGoogle Scholar
  8. Goodfellow, I., Bengio, Y. and Courvile, A. (2016). Deep Learning. MIT Press. Cambridge, Massachusetts, USA.zbMATHGoogle Scholar
  9. Holland, H. J. (1992). Genetic algorithm. Scientific American 267, 1, 66–73.CrossRefGoogle Scholar
  10. Jayalakshmi, T. and Santhakumaran, A. (2011). Statistical normalization and back propagation for classification. Int. J. Computer Theory and Engineering 3, 1, 89–93.CrossRefGoogle Scholar
  11. Kasa, I. (1976). A circle fitting procedure and its error analysis. IEEE Trans. Instrumentation and Measurement IM-25, 1, 8–14.CrossRefGoogle Scholar
  12. Kim, J. H. (2014). A Numerical Study on the Effect of Material Properties for the Surface Deflection in Rectangular Embossing Process. M. S. Thesis. Seoul National University of Science and Technology. Seoul, Korea.Google Scholar
  13. Kingma, P. D. and Ba, L. J. (2015). ADAM: A method for stochastic optimization. Proc. 3rd Int. Conf. Learning Representations (ICLR), San Diego, California, USA.Google Scholar
  14. Liew, K. M., Tan, H., Ray, T. and Tan, M. J. (2004). Optimal process design of sheet metal forming for minimum springback via an integrated neural network evolutionary algorithm. Structural and Multidisciplinary Optimization 26, 3–4, 284–294.CrossRefGoogle Scholar
  15. Liu, W., Liu, Q., Ruan, F., Liang, Z. and Qiu, H. (2007). Springback prediction for sheet metal forming based on GA-ANN technology. J. Materials Processing Technology, 187–188, 227–231.CrossRefGoogle Scholar
  16. Mckay, M. D. (1979). Comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics 21, 2, 239–245.MathSciNetzbMATHGoogle Scholar
  17. Park, C. D., Chung, W. J. and Kim, B. M. (2007). A numerical and experimental study of surface deflections in automobile exterior panels. J. Materials Processing Technology, 187–188, 99–102.CrossRefGoogle Scholar
  18. Son, D. K. and Chung, W. J. (2019). Prediction model of surface deflection rectangular drawing products using finite element analysis and machine learning. Trans. Korean Society of Mechanical Engineers 43, 1, 41–50.CrossRefGoogle Scholar
  19. Xie, Y., Tang, W., Zhang, F., Pan, B., Yue, Y. and Feng, M. (2019). Optimization of variable blank holder force based on a sharing niching RBF neural network and an improved NSGA-II algorithm. Int. J. Precision Engineering and Manufacturing 20, 2, 285–229.CrossRefGoogle Scholar
  20. Yoshida, K. (1993). Handbook of Ease or Difficulty in Press Forming. The Thin Sheet Steel Forming Technology Study Association. Japan.Google Scholar

Copyright information

© KSAE/112-08 2019

Authors and Affiliations

  1. 1.Department of Mechanical Design and Manufacturing Engineering, Graduate SchoolSeoul National University of Science and TechnologySeoulKorea
  2. 2.Department of Mechanical System Design EngineeringSeoul National University of Science and TechnologySeoulKorea

Personalised recommendations